{ "metadata": { "total_size": 14806597632 }, "weight_map": { "layers.0.attention.dyn_w_proj.dd": "pytorch_model-00001-of-00003.bin", "layers.0.attention.dyn_w_proj.dw1": "pytorch_model-00001-of-00003.bin", "layers.0.attention.dyn_w_proj.dw_m": "pytorch_model-00001-of-00003.bin", "layers.0.attention.dyn_w_proj.post_proj.w": "pytorch_model-00001-of-00003.bin", "layers.0.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00001-of-00003.bin", "layers.0.attention.dyn_w_proj.qkw": "pytorch_model-00001-of-00003.bin", "layers.0.attention.dyn_w_proj.qkw_m": "pytorch_model-00001-of-00003.bin", "layers.0.attention.k_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.0.attention.kv_cache.k_cache": "pytorch_model-00001-of-00003.bin", "layers.0.attention.kv_cache.kw_cache": "pytorch_model-00001-of-00003.bin", "layers.0.attention.kv_cache.v_cache": "pytorch_model-00001-of-00003.bin", "layers.0.attention.q_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.0.attention.wo.bias": "pytorch_model-00001-of-00003.bin", "layers.0.attention.wo.weight": "pytorch_model-00001-of-00003.bin", "layers.0.attention.wqkv.bias": "pytorch_model-00001-of-00003.bin", "layers.0.attention.wqkv.weight": "pytorch_model-00001-of-00003.bin", "layers.0.attention_norm.bias": "pytorch_model-00001-of-00003.bin", "layers.0.attention_norm.weight": "pytorch_model-00001-of-00003.bin", "layers.0.feed_forward.w1.bias": "pytorch_model-00001-of-00003.bin", "layers.0.feed_forward.w1.weight": "pytorch_model-00001-of-00003.bin", "layers.0.feed_forward.w2.bias": "pytorch_model-00001-of-00003.bin", "layers.0.feed_forward.w2.weight": "pytorch_model-00001-of-00003.bin", "layers.0.ffn_norm.bias": "pytorch_model-00001-of-00003.bin", "layers.0.ffn_norm.weight": "pytorch_model-00001-of-00003.bin", "layers.1.attention.dyn_w_proj.dd": "pytorch_model-00001-of-00003.bin", "layers.1.attention.dyn_w_proj.dw1": "pytorch_model-00001-of-00003.bin", "layers.1.attention.dyn_w_proj.dw_m": "pytorch_model-00001-of-00003.bin", "layers.1.attention.dyn_w_proj.post_proj.w": "pytorch_model-00001-of-00003.bin", "layers.1.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00001-of-00003.bin", "layers.1.attention.dyn_w_proj.qkw": "pytorch_model-00001-of-00003.bin", "layers.1.attention.dyn_w_proj.qkw_m": "pytorch_model-00001-of-00003.bin", "layers.1.attention.k_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.1.attention.kv_cache.k_cache": "pytorch_model-00001-of-00003.bin", "layers.1.attention.kv_cache.kw_cache": "pytorch_model-00001-of-00003.bin", "layers.1.attention.kv_cache.v_cache": "pytorch_model-00001-of-00003.bin", "layers.1.attention.q_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.1.attention.wo.bias": "pytorch_model-00001-of-00003.bin", "layers.1.attention.wo.weight": "pytorch_model-00001-of-00003.bin", "layers.1.attention.wqkv.bias": "pytorch_model-00001-of-00003.bin", "layers.1.attention.wqkv.weight": "pytorch_model-00001-of-00003.bin", "layers.1.attention_norm.bias": "pytorch_model-00001-of-00003.bin", "layers.1.attention_norm.weight": "pytorch_model-00001-of-00003.bin", "layers.1.feed_forward.w1.bias": "pytorch_model-00001-of-00003.bin", "layers.1.feed_forward.w1.weight": "pytorch_model-00001-of-00003.bin", "layers.1.feed_forward.w2.bias": "pytorch_model-00001-of-00003.bin", "layers.1.feed_forward.w2.weight": "pytorch_model-00001-of-00003.bin", "layers.1.ffn_norm.bias": "pytorch_model-00001-of-00003.bin", "layers.1.ffn_norm.weight": "pytorch_model-00001-of-00003.bin", "layers.10.attention.dyn_w_proj.dd": "pytorch_model-00001-of-00003.bin", "layers.10.attention.dyn_w_proj.dw1": "pytorch_model-00001-of-00003.bin", "layers.10.attention.dyn_w_proj.dw_m": "pytorch_model-00001-of-00003.bin", "layers.10.attention.dyn_w_proj.post_proj.w": "pytorch_model-00001-of-00003.bin", "layers.10.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00001-of-00003.bin", "layers.10.attention.dyn_w_proj.qkw": "pytorch_model-00001-of-00003.bin", "layers.10.attention.dyn_w_proj.qkw_m": "pytorch_model-00001-of-00003.bin", "layers.10.attention.k_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.10.attention.kv_cache.k_cache": "pytorch_model-00001-of-00003.bin", "layers.10.attention.kv_cache.kw_cache": "pytorch_model-00001-of-00003.bin", "layers.10.attention.kv_cache.v_cache": "pytorch_model-00001-of-00003.bin", "layers.10.attention.q_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.10.attention.wo.bias": "pytorch_model-00001-of-00003.bin", "layers.10.attention.wo.weight": "pytorch_model-00001-of-00003.bin", "layers.10.attention.wqkv.bias": "pytorch_model-00001-of-00003.bin", "layers.10.attention.wqkv.weight": "pytorch_model-00001-of-00003.bin", "layers.10.attention_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.10.attention_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.10.feed_forward.w1.bias": "pytorch_model-00002-of-00003.bin", "layers.10.feed_forward.w1.weight": "pytorch_model-00002-of-00003.bin", "layers.10.feed_forward.w2.bias": "pytorch_model-00002-of-00003.bin", "layers.10.feed_forward.w2.weight": "pytorch_model-00002-of-00003.bin", "layers.10.ffn_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.10.ffn_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.11.attention.dyn_w_proj.dd": "pytorch_model-00002-of-00003.bin", "layers.11.attention.dyn_w_proj.dw1": "pytorch_model-00002-of-00003.bin", "layers.11.attention.dyn_w_proj.dw_m": "pytorch_model-00002-of-00003.bin", "layers.11.attention.dyn_w_proj.post_proj.w": "pytorch_model-00002-of-00003.bin", "layers.11.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00002-of-00003.bin", "layers.11.attention.dyn_w_proj.qkw": "pytorch_model-00002-of-00003.bin", "layers.11.attention.dyn_w_proj.qkw_m": "pytorch_model-00002-of-00003.bin", "layers.11.attention.k_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.11.attention.kv_cache.k_cache": "pytorch_model-00002-of-00003.bin", "layers.11.attention.kv_cache.kw_cache": "pytorch_model-00002-of-00003.bin", "layers.11.attention.kv_cache.v_cache": "pytorch_model-00002-of-00003.bin", "layers.11.attention.q_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.11.attention.wo.bias": "pytorch_model-00002-of-00003.bin", "layers.11.attention.wo.weight": "pytorch_model-00002-of-00003.bin", "layers.11.attention.wqkv.bias": "pytorch_model-00002-of-00003.bin", "layers.11.attention.wqkv.weight": "pytorch_model-00002-of-00003.bin", "layers.11.attention_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.11.attention_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.11.feed_forward.w1.bias": "pytorch_model-00002-of-00003.bin", "layers.11.feed_forward.w1.weight": "pytorch_model-00002-of-00003.bin", "layers.11.feed_forward.w2.bias": "pytorch_model-00002-of-00003.bin", "layers.11.feed_forward.w2.weight": "pytorch_model-00002-of-00003.bin", "layers.11.ffn_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.11.ffn_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.12.attention.dyn_w_proj.dd": "pytorch_model-00002-of-00003.bin", "layers.12.attention.dyn_w_proj.dw1": "pytorch_model-00002-of-00003.bin", "layers.12.attention.dyn_w_proj.dw_m": "pytorch_model-00002-of-00003.bin", "layers.12.attention.dyn_w_proj.post_proj.w": "pytorch_model-00002-of-00003.bin", "layers.12.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00002-of-00003.bin", "layers.12.attention.dyn_w_proj.qkw": "pytorch_model-00002-of-00003.bin", "layers.12.attention.dyn_w_proj.qkw_m": "pytorch_model-00002-of-00003.bin", "layers.12.attention.k_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.12.attention.kv_cache.k_cache": "pytorch_model-00002-of-00003.bin", "layers.12.attention.kv_cache.kw_cache": "pytorch_model-00002-of-00003.bin", "layers.12.attention.kv_cache.v_cache": "pytorch_model-00002-of-00003.bin", "layers.12.attention.q_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.12.attention.wo.bias": "pytorch_model-00002-of-00003.bin", "layers.12.attention.wo.weight": "pytorch_model-00002-of-00003.bin", "layers.12.attention.wqkv.bias": "pytorch_model-00002-of-00003.bin", "layers.12.attention.wqkv.weight": "pytorch_model-00002-of-00003.bin", "layers.12.attention_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.12.attention_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.12.feed_forward.w1.bias": "pytorch_model-00002-of-00003.bin", "layers.12.feed_forward.w1.weight": "pytorch_model-00002-of-00003.bin", "layers.12.feed_forward.w2.bias": "pytorch_model-00002-of-00003.bin", "layers.12.feed_forward.w2.weight": "pytorch_model-00002-of-00003.bin", "layers.12.ffn_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.12.ffn_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.13.attention.dyn_w_proj.dd": "pytorch_model-00002-of-00003.bin", "layers.13.attention.dyn_w_proj.dw1": "pytorch_model-00002-of-00003.bin", "layers.13.attention.dyn_w_proj.dw_m": "pytorch_model-00002-of-00003.bin", "layers.13.attention.dyn_w_proj.post_proj.w": "pytorch_model-00002-of-00003.bin", "layers.13.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00002-of-00003.bin", "layers.13.attention.dyn_w_proj.qkw": "pytorch_model-00002-of-00003.bin", "layers.13.attention.dyn_w_proj.qkw_m": "pytorch_model-00002-of-00003.bin", "layers.13.attention.k_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.13.attention.kv_cache.k_cache": "pytorch_model-00002-of-00003.bin", "layers.13.attention.kv_cache.kw_cache": "pytorch_model-00002-of-00003.bin", "layers.13.attention.kv_cache.v_cache": "pytorch_model-00002-of-00003.bin", "layers.13.attention.q_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.13.attention.wo.bias": "pytorch_model-00002-of-00003.bin", "layers.13.attention.wo.weight": "pytorch_model-00002-of-00003.bin", "layers.13.attention.wqkv.bias": "pytorch_model-00002-of-00003.bin", "layers.13.attention.wqkv.weight": "pytorch_model-00002-of-00003.bin", "layers.13.attention_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.13.attention_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.13.feed_forward.w1.bias": "pytorch_model-00002-of-00003.bin", "layers.13.feed_forward.w1.weight": "pytorch_model-00002-of-00003.bin", "layers.13.feed_forward.w2.bias": "pytorch_model-00002-of-00003.bin", "layers.13.feed_forward.w2.weight": "pytorch_model-00002-of-00003.bin", "layers.13.ffn_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.13.ffn_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.14.attention.dyn_w_proj.dd": "pytorch_model-00002-of-00003.bin", "layers.14.attention.dyn_w_proj.dw1": "pytorch_model-00002-of-00003.bin", "layers.14.attention.dyn_w_proj.dw_m": "pytorch_model-00002-of-00003.bin", "layers.14.attention.dyn_w_proj.post_proj.w": "pytorch_model-00002-of-00003.bin", "layers.14.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00002-of-00003.bin", "layers.14.attention.dyn_w_proj.qkw": "pytorch_model-00002-of-00003.bin", "layers.14.attention.dyn_w_proj.qkw_m": "pytorch_model-00002-of-00003.bin", "layers.14.attention.k_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.14.attention.kv_cache.k_cache": "pytorch_model-00002-of-00003.bin", "layers.14.attention.kv_cache.kw_cache": "pytorch_model-00002-of-00003.bin", "layers.14.attention.kv_cache.v_cache": "pytorch_model-00002-of-00003.bin", "layers.14.attention.q_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.14.attention.wo.bias": "pytorch_model-00002-of-00003.bin", "layers.14.attention.wo.weight": "pytorch_model-00002-of-00003.bin", "layers.14.attention.wqkv.bias": "pytorch_model-00002-of-00003.bin", "layers.14.attention.wqkv.weight": "pytorch_model-00002-of-00003.bin", "layers.14.attention_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.14.attention_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.14.feed_forward.w1.bias": "pytorch_model-00002-of-00003.bin", "layers.14.feed_forward.w1.weight": "pytorch_model-00002-of-00003.bin", "layers.14.feed_forward.w2.bias": "pytorch_model-00002-of-00003.bin", "layers.14.feed_forward.w2.weight": "pytorch_model-00002-of-00003.bin", "layers.14.ffn_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.14.ffn_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.15.attention.dyn_w_proj.dd": "pytorch_model-00002-of-00003.bin", "layers.15.attention.dyn_w_proj.dw1": "pytorch_model-00002-of-00003.bin", "layers.15.attention.dyn_w_proj.dw_m": "pytorch_model-00002-of-00003.bin", "layers.15.attention.dyn_w_proj.post_proj.w": "pytorch_model-00002-of-00003.bin", "layers.15.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00002-of-00003.bin", "layers.15.attention.dyn_w_proj.qkw": "pytorch_model-00002-of-00003.bin", "layers.15.attention.dyn_w_proj.qkw_m": "pytorch_model-00002-of-00003.bin", "layers.15.attention.k_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.15.attention.kv_cache.k_cache": "pytorch_model-00002-of-00003.bin", "layers.15.attention.kv_cache.kw_cache": "pytorch_model-00002-of-00003.bin", "layers.15.attention.kv_cache.v_cache": "pytorch_model-00002-of-00003.bin", "layers.15.attention.q_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.15.attention.wo.bias": "pytorch_model-00002-of-00003.bin", "layers.15.attention.wo.weight": "pytorch_model-00002-of-00003.bin", "layers.15.attention.wqkv.bias": "pytorch_model-00002-of-00003.bin", "layers.15.attention.wqkv.weight": "pytorch_model-00002-of-00003.bin", "layers.15.attention_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.15.attention_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.15.feed_forward.w1.bias": "pytorch_model-00002-of-00003.bin", "layers.15.feed_forward.w1.weight": "pytorch_model-00002-of-00003.bin", "layers.15.feed_forward.w2.bias": "pytorch_model-00002-of-00003.bin", "layers.15.feed_forward.w2.weight": "pytorch_model-00002-of-00003.bin", "layers.15.ffn_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.15.ffn_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.16.attention.dyn_w_proj.dd": "pytorch_model-00002-of-00003.bin", "layers.16.attention.dyn_w_proj.dw1": "pytorch_model-00002-of-00003.bin", "layers.16.attention.dyn_w_proj.dw_m": "pytorch_model-00002-of-00003.bin", "layers.16.attention.dyn_w_proj.post_proj.w": "pytorch_model-00002-of-00003.bin", "layers.16.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00002-of-00003.bin", "layers.16.attention.dyn_w_proj.qkw": "pytorch_model-00002-of-00003.bin", "layers.16.attention.dyn_w_proj.qkw_m": "pytorch_model-00002-of-00003.bin", "layers.16.attention.k_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.16.attention.kv_cache.k_cache": "pytorch_model-00002-of-00003.bin", "layers.16.attention.kv_cache.kw_cache": "pytorch_model-00002-of-00003.bin", "layers.16.attention.kv_cache.v_cache": "pytorch_model-00002-of-00003.bin", "layers.16.attention.q_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.16.attention.wo.bias": "pytorch_model-00002-of-00003.bin", "layers.16.attention.wo.weight": "pytorch_model-00002-of-00003.bin", "layers.16.attention.wqkv.bias": "pytorch_model-00002-of-00003.bin", "layers.16.attention.wqkv.weight": "pytorch_model-00002-of-00003.bin", "layers.16.attention_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.16.attention_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.16.feed_forward.w1.bias": "pytorch_model-00002-of-00003.bin", "layers.16.feed_forward.w1.weight": "pytorch_model-00002-of-00003.bin", "layers.16.feed_forward.w2.bias": "pytorch_model-00002-of-00003.bin", "layers.16.feed_forward.w2.weight": "pytorch_model-00002-of-00003.bin", "layers.16.ffn_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.16.ffn_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.17.attention.dyn_w_proj.dd": "pytorch_model-00002-of-00003.bin", "layers.17.attention.dyn_w_proj.dw1": "pytorch_model-00002-of-00003.bin", "layers.17.attention.dyn_w_proj.dw_m": "pytorch_model-00002-of-00003.bin", "layers.17.attention.dyn_w_proj.post_proj.w": "pytorch_model-00002-of-00003.bin", "layers.17.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00002-of-00003.bin", "layers.17.attention.dyn_w_proj.qkw": "pytorch_model-00002-of-00003.bin", "layers.17.attention.dyn_w_proj.qkw_m": "pytorch_model-00002-of-00003.bin", "layers.17.attention.k_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.17.attention.kv_cache.k_cache": "pytorch_model-00002-of-00003.bin", "layers.17.attention.kv_cache.kw_cache": "pytorch_model-00002-of-00003.bin", "layers.17.attention.kv_cache.v_cache": "pytorch_model-00002-of-00003.bin", "layers.17.attention.q_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.17.attention.wo.bias": "pytorch_model-00002-of-00003.bin", "layers.17.attention.wo.weight": "pytorch_model-00002-of-00003.bin", "layers.17.attention.wqkv.bias": "pytorch_model-00002-of-00003.bin", "layers.17.attention.wqkv.weight": "pytorch_model-00002-of-00003.bin", "layers.17.attention_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.17.attention_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.17.feed_forward.w1.bias": "pytorch_model-00002-of-00003.bin", "layers.17.feed_forward.w1.weight": "pytorch_model-00002-of-00003.bin", "layers.17.feed_forward.w2.bias": "pytorch_model-00002-of-00003.bin", "layers.17.feed_forward.w2.weight": "pytorch_model-00002-of-00003.bin", "layers.17.ffn_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.17.ffn_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.18.attention.dyn_w_proj.dd": "pytorch_model-00002-of-00003.bin", "layers.18.attention.dyn_w_proj.dw1": "pytorch_model-00002-of-00003.bin", "layers.18.attention.dyn_w_proj.dw_m": "pytorch_model-00002-of-00003.bin", "layers.18.attention.dyn_w_proj.post_proj.w": "pytorch_model-00002-of-00003.bin", "layers.18.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00002-of-00003.bin", "layers.18.attention.dyn_w_proj.qkw": "pytorch_model-00002-of-00003.bin", "layers.18.attention.dyn_w_proj.qkw_m": "pytorch_model-00002-of-00003.bin", "layers.18.attention.k_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.18.attention.kv_cache.k_cache": "pytorch_model-00002-of-00003.bin", "layers.18.attention.kv_cache.kw_cache": "pytorch_model-00002-of-00003.bin", "layers.18.attention.kv_cache.v_cache": "pytorch_model-00002-of-00003.bin", "layers.18.attention.q_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.18.attention.wo.bias": "pytorch_model-00002-of-00003.bin", "layers.18.attention.wo.weight": "pytorch_model-00002-of-00003.bin", "layers.18.attention.wqkv.bias": "pytorch_model-00002-of-00003.bin", "layers.18.attention.wqkv.weight": "pytorch_model-00002-of-00003.bin", "layers.18.attention_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.18.attention_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.18.feed_forward.w1.bias": "pytorch_model-00002-of-00003.bin", "layers.18.feed_forward.w1.weight": "pytorch_model-00002-of-00003.bin", "layers.18.feed_forward.w2.bias": "pytorch_model-00002-of-00003.bin", "layers.18.feed_forward.w2.weight": "pytorch_model-00002-of-00003.bin", "layers.18.ffn_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.18.ffn_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.19.attention.dyn_w_proj.dd": "pytorch_model-00002-of-00003.bin", "layers.19.attention.dyn_w_proj.dw1": "pytorch_model-00002-of-00003.bin", "layers.19.attention.dyn_w_proj.dw_m": "pytorch_model-00002-of-00003.bin", "layers.19.attention.dyn_w_proj.post_proj.w": "pytorch_model-00002-of-00003.bin", "layers.19.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00002-of-00003.bin", "layers.19.attention.dyn_w_proj.qkw": "pytorch_model-00002-of-00003.bin", "layers.19.attention.dyn_w_proj.qkw_m": "pytorch_model-00002-of-00003.bin", "layers.19.attention.k_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.19.attention.kv_cache.k_cache": "pytorch_model-00002-of-00003.bin", "layers.19.attention.kv_cache.kw_cache": "pytorch_model-00002-of-00003.bin", "layers.19.attention.kv_cache.v_cache": "pytorch_model-00002-of-00003.bin", "layers.19.attention.q_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.19.attention.wo.bias": "pytorch_model-00002-of-00003.bin", "layers.19.attention.wo.weight": "pytorch_model-00002-of-00003.bin", "layers.19.attention.wqkv.bias": "pytorch_model-00002-of-00003.bin", "layers.19.attention.wqkv.weight": "pytorch_model-00002-of-00003.bin", "layers.19.attention_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.19.attention_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.19.feed_forward.w1.bias": "pytorch_model-00002-of-00003.bin", "layers.19.feed_forward.w1.weight": "pytorch_model-00002-of-00003.bin", "layers.19.feed_forward.w2.bias": "pytorch_model-00002-of-00003.bin", "layers.19.feed_forward.w2.weight": "pytorch_model-00002-of-00003.bin", "layers.19.ffn_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.19.ffn_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.2.attention.dyn_w_proj.dd": "pytorch_model-00001-of-00003.bin", "layers.2.attention.dyn_w_proj.dw1": "pytorch_model-00001-of-00003.bin", "layers.2.attention.dyn_w_proj.dw_m": "pytorch_model-00001-of-00003.bin", "layers.2.attention.dyn_w_proj.post_proj.w": "pytorch_model-00001-of-00003.bin", "layers.2.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00001-of-00003.bin", "layers.2.attention.dyn_w_proj.qkw": "pytorch_model-00001-of-00003.bin", "layers.2.attention.dyn_w_proj.qkw_m": "pytorch_model-00001-of-00003.bin", "layers.2.attention.k_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.2.attention.kv_cache.k_cache": "pytorch_model-00001-of-00003.bin", "layers.2.attention.kv_cache.kw_cache": "pytorch_model-00001-of-00003.bin", "layers.2.attention.kv_cache.v_cache": "pytorch_model-00001-of-00003.bin", "layers.2.attention.q_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.2.attention.wo.bias": "pytorch_model-00001-of-00003.bin", "layers.2.attention.wo.weight": "pytorch_model-00001-of-00003.bin", "layers.2.attention.wqkv.bias": "pytorch_model-00001-of-00003.bin", "layers.2.attention.wqkv.weight": "pytorch_model-00001-of-00003.bin", "layers.2.attention_norm.bias": "pytorch_model-00001-of-00003.bin", "layers.2.attention_norm.weight": "pytorch_model-00001-of-00003.bin", "layers.2.feed_forward.w1.bias": "pytorch_model-00001-of-00003.bin", "layers.2.feed_forward.w1.weight": "pytorch_model-00001-of-00003.bin", "layers.2.feed_forward.w2.bias": "pytorch_model-00001-of-00003.bin", "layers.2.feed_forward.w2.weight": "pytorch_model-00001-of-00003.bin", "layers.2.ffn_norm.bias": "pytorch_model-00001-of-00003.bin", "layers.2.ffn_norm.weight": "pytorch_model-00001-of-00003.bin", "layers.20.attention.dyn_w_proj.dd": "pytorch_model-00002-of-00003.bin", "layers.20.attention.dyn_w_proj.dw1": "pytorch_model-00002-of-00003.bin", "layers.20.attention.dyn_w_proj.dw_m": "pytorch_model-00002-of-00003.bin", "layers.20.attention.dyn_w_proj.post_proj.w": "pytorch_model-00002-of-00003.bin", "layers.20.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00002-of-00003.bin", "layers.20.attention.dyn_w_proj.qkw": "pytorch_model-00002-of-00003.bin", "layers.20.attention.dyn_w_proj.qkw_m": "pytorch_model-00002-of-00003.bin", "layers.20.attention.k_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.20.attention.kv_cache.k_cache": "pytorch_model-00002-of-00003.bin", "layers.20.attention.kv_cache.kw_cache": "pytorch_model-00002-of-00003.bin", "layers.20.attention.kv_cache.v_cache": "pytorch_model-00002-of-00003.bin", "layers.20.attention.q_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.20.attention.wo.bias": "pytorch_model-00002-of-00003.bin", "layers.20.attention.wo.weight": "pytorch_model-00002-of-00003.bin", "layers.20.attention.wqkv.bias": "pytorch_model-00002-of-00003.bin", "layers.20.attention.wqkv.weight": "pytorch_model-00002-of-00003.bin", "layers.20.attention_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.20.attention_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.20.feed_forward.w1.bias": "pytorch_model-00002-of-00003.bin", "layers.20.feed_forward.w1.weight": "pytorch_model-00002-of-00003.bin", "layers.20.feed_forward.w2.bias": "pytorch_model-00002-of-00003.bin", "layers.20.feed_forward.w2.weight": "pytorch_model-00002-of-00003.bin", "layers.20.ffn_norm.bias": "pytorch_model-00002-of-00003.bin", "layers.20.ffn_norm.weight": "pytorch_model-00002-of-00003.bin", "layers.21.attention.dyn_w_proj.dd": "pytorch_model-00002-of-00003.bin", "layers.21.attention.dyn_w_proj.dw1": "pytorch_model-00002-of-00003.bin", "layers.21.attention.dyn_w_proj.dw_m": "pytorch_model-00002-of-00003.bin", "layers.21.attention.dyn_w_proj.post_proj.w": "pytorch_model-00002-of-00003.bin", "layers.21.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00002-of-00003.bin", "layers.21.attention.dyn_w_proj.qkw": "pytorch_model-00002-of-00003.bin", "layers.21.attention.dyn_w_proj.qkw_m": "pytorch_model-00002-of-00003.bin", "layers.21.attention.k_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.21.attention.kv_cache.k_cache": "pytorch_model-00002-of-00003.bin", "layers.21.attention.kv_cache.kw_cache": "pytorch_model-00002-of-00003.bin", "layers.21.attention.kv_cache.v_cache": "pytorch_model-00002-of-00003.bin", "layers.21.attention.q_norm.scale": "pytorch_model-00002-of-00003.bin", "layers.21.attention.wo.bias": "pytorch_model-00002-of-00003.bin", "layers.21.attention.wo.weight": "pytorch_model-00002-of-00003.bin", "layers.21.attention.wqkv.bias": "pytorch_model-00002-of-00003.bin", "layers.21.attention.wqkv.weight": "pytorch_model-00002-of-00003.bin", "layers.21.attention_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.21.attention_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.21.feed_forward.w1.bias": "pytorch_model-00002-of-00003.bin", "layers.21.feed_forward.w1.weight": "pytorch_model-00002-of-00003.bin", "layers.21.feed_forward.w2.bias": "pytorch_model-00003-of-00003.bin", "layers.21.feed_forward.w2.weight": "pytorch_model-00003-of-00003.bin", "layers.21.ffn_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.21.ffn_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.22.attention.dyn_w_proj.dd": "pytorch_model-00003-of-00003.bin", "layers.22.attention.dyn_w_proj.dw1": "pytorch_model-00003-of-00003.bin", "layers.22.attention.dyn_w_proj.dw_m": "pytorch_model-00003-of-00003.bin", "layers.22.attention.dyn_w_proj.post_proj.w": "pytorch_model-00003-of-00003.bin", "layers.22.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00003-of-00003.bin", "layers.22.attention.dyn_w_proj.qkw": "pytorch_model-00003-of-00003.bin", "layers.22.attention.dyn_w_proj.qkw_m": "pytorch_model-00003-of-00003.bin", "layers.22.attention.k_norm.scale": "pytorch_model-00003-of-00003.bin", "layers.22.attention.kv_cache.k_cache": "pytorch_model-00003-of-00003.bin", "layers.22.attention.kv_cache.kw_cache": "pytorch_model-00003-of-00003.bin", "layers.22.attention.kv_cache.v_cache": "pytorch_model-00003-of-00003.bin", "layers.22.attention.q_norm.scale": "pytorch_model-00003-of-00003.bin", "layers.22.attention.wo.bias": "pytorch_model-00003-of-00003.bin", "layers.22.attention.wo.weight": "pytorch_model-00003-of-00003.bin", "layers.22.attention.wqkv.bias": "pytorch_model-00003-of-00003.bin", "layers.22.attention.wqkv.weight": "pytorch_model-00003-of-00003.bin", "layers.22.attention_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.22.attention_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.22.feed_forward.w1.bias": "pytorch_model-00003-of-00003.bin", "layers.22.feed_forward.w1.weight": "pytorch_model-00003-of-00003.bin", "layers.22.feed_forward.w2.bias": "pytorch_model-00003-of-00003.bin", "layers.22.feed_forward.w2.weight": "pytorch_model-00003-of-00003.bin", "layers.22.ffn_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.22.ffn_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.23.attention.dyn_w_proj.dd": "pytorch_model-00003-of-00003.bin", "layers.23.attention.dyn_w_proj.dw1": "pytorch_model-00003-of-00003.bin", "layers.23.attention.dyn_w_proj.dw_m": "pytorch_model-00003-of-00003.bin", "layers.23.attention.dyn_w_proj.post_proj.w": "pytorch_model-00003-of-00003.bin", "layers.23.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00003-of-00003.bin", "layers.23.attention.dyn_w_proj.qkw": "pytorch_model-00003-of-00003.bin", "layers.23.attention.dyn_w_proj.qkw_m": "pytorch_model-00003-of-00003.bin", "layers.23.attention.k_norm.scale": "pytorch_model-00003-of-00003.bin", "layers.23.attention.kv_cache.k_cache": "pytorch_model-00003-of-00003.bin", "layers.23.attention.kv_cache.kw_cache": "pytorch_model-00003-of-00003.bin", "layers.23.attention.kv_cache.v_cache": "pytorch_model-00003-of-00003.bin", "layers.23.attention.q_norm.scale": "pytorch_model-00003-of-00003.bin", "layers.23.attention.wo.bias": "pytorch_model-00003-of-00003.bin", "layers.23.attention.wo.weight": "pytorch_model-00003-of-00003.bin", "layers.23.attention.wqkv.bias": "pytorch_model-00003-of-00003.bin", "layers.23.attention.wqkv.weight": "pytorch_model-00003-of-00003.bin", "layers.23.attention_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.23.attention_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.23.feed_forward.w1.bias": "pytorch_model-00003-of-00003.bin", "layers.23.feed_forward.w1.weight": "pytorch_model-00003-of-00003.bin", "layers.23.feed_forward.w2.bias": "pytorch_model-00003-of-00003.bin", "layers.23.feed_forward.w2.weight": "pytorch_model-00003-of-00003.bin", "layers.23.ffn_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.23.ffn_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.24.attention.dyn_w_proj.dd": "pytorch_model-00003-of-00003.bin", "layers.24.attention.dyn_w_proj.dw1": "pytorch_model-00003-of-00003.bin", "layers.24.attention.dyn_w_proj.dw_m": "pytorch_model-00003-of-00003.bin", "layers.24.attention.dyn_w_proj.post_proj.w": "pytorch_model-00003-of-00003.bin", "layers.24.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00003-of-00003.bin", "layers.24.attention.dyn_w_proj.qkw": "pytorch_model-00003-of-00003.bin", "layers.24.attention.dyn_w_proj.qkw_m": "pytorch_model-00003-of-00003.bin", "layers.24.attention.k_norm.scale": "pytorch_model-00003-of-00003.bin", "layers.24.attention.kv_cache.k_cache": "pytorch_model-00003-of-00003.bin", "layers.24.attention.kv_cache.kw_cache": "pytorch_model-00003-of-00003.bin", "layers.24.attention.kv_cache.v_cache": "pytorch_model-00003-of-00003.bin", "layers.24.attention.q_norm.scale": "pytorch_model-00003-of-00003.bin", "layers.24.attention.wo.bias": "pytorch_model-00003-of-00003.bin", "layers.24.attention.wo.weight": "pytorch_model-00003-of-00003.bin", "layers.24.attention.wqkv.bias": "pytorch_model-00003-of-00003.bin", "layers.24.attention.wqkv.weight": "pytorch_model-00003-of-00003.bin", "layers.24.attention_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.24.attention_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.24.feed_forward.w1.bias": "pytorch_model-00003-of-00003.bin", "layers.24.feed_forward.w1.weight": "pytorch_model-00003-of-00003.bin", "layers.24.feed_forward.w2.bias": "pytorch_model-00003-of-00003.bin", "layers.24.feed_forward.w2.weight": "pytorch_model-00003-of-00003.bin", "layers.24.ffn_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.24.ffn_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.25.attention.dyn_w_proj.dd": "pytorch_model-00003-of-00003.bin", "layers.25.attention.dyn_w_proj.dw1": "pytorch_model-00003-of-00003.bin", "layers.25.attention.dyn_w_proj.dw_m": "pytorch_model-00003-of-00003.bin", "layers.25.attention.dyn_w_proj.post_proj.w": "pytorch_model-00003-of-00003.bin", "layers.25.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00003-of-00003.bin", "layers.25.attention.dyn_w_proj.qkw": "pytorch_model-00003-of-00003.bin", "layers.25.attention.dyn_w_proj.qkw_m": "pytorch_model-00003-of-00003.bin", "layers.25.attention.k_norm.scale": "pytorch_model-00003-of-00003.bin", "layers.25.attention.kv_cache.k_cache": "pytorch_model-00003-of-00003.bin", "layers.25.attention.kv_cache.kw_cache": "pytorch_model-00003-of-00003.bin", "layers.25.attention.kv_cache.v_cache": "pytorch_model-00003-of-00003.bin", "layers.25.attention.q_norm.scale": "pytorch_model-00003-of-00003.bin", "layers.25.attention.wo.bias": "pytorch_model-00003-of-00003.bin", "layers.25.attention.wo.weight": "pytorch_model-00003-of-00003.bin", "layers.25.attention.wqkv.bias": "pytorch_model-00003-of-00003.bin", "layers.25.attention.wqkv.weight": "pytorch_model-00003-of-00003.bin", "layers.25.attention_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.25.attention_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.25.feed_forward.w1.bias": "pytorch_model-00003-of-00003.bin", "layers.25.feed_forward.w1.weight": "pytorch_model-00003-of-00003.bin", "layers.25.feed_forward.w2.bias": "pytorch_model-00003-of-00003.bin", "layers.25.feed_forward.w2.weight": "pytorch_model-00003-of-00003.bin", "layers.25.ffn_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.25.ffn_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.26.attention.dyn_w_proj.dd": "pytorch_model-00003-of-00003.bin", "layers.26.attention.dyn_w_proj.dw1": "pytorch_model-00003-of-00003.bin", "layers.26.attention.dyn_w_proj.dw_m": "pytorch_model-00003-of-00003.bin", "layers.26.attention.dyn_w_proj.post_proj.w": "pytorch_model-00003-of-00003.bin", "layers.26.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00003-of-00003.bin", "layers.26.attention.dyn_w_proj.qkw": "pytorch_model-00003-of-00003.bin", "layers.26.attention.dyn_w_proj.qkw_m": "pytorch_model-00003-of-00003.bin", "layers.26.attention.k_norm.scale": "pytorch_model-00003-of-00003.bin", "layers.26.attention.kv_cache.k_cache": "pytorch_model-00003-of-00003.bin", "layers.26.attention.kv_cache.kw_cache": "pytorch_model-00003-of-00003.bin", "layers.26.attention.kv_cache.v_cache": "pytorch_model-00003-of-00003.bin", "layers.26.attention.q_norm.scale": "pytorch_model-00003-of-00003.bin", "layers.26.attention.wo.bias": "pytorch_model-00003-of-00003.bin", "layers.26.attention.wo.weight": "pytorch_model-00003-of-00003.bin", "layers.26.attention.wqkv.bias": "pytorch_model-00003-of-00003.bin", "layers.26.attention.wqkv.weight": "pytorch_model-00003-of-00003.bin", "layers.26.attention_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.26.attention_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.26.feed_forward.w1.bias": "pytorch_model-00003-of-00003.bin", "layers.26.feed_forward.w1.weight": "pytorch_model-00003-of-00003.bin", "layers.26.feed_forward.w2.bias": "pytorch_model-00003-of-00003.bin", "layers.26.feed_forward.w2.weight": "pytorch_model-00003-of-00003.bin", "layers.26.ffn_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.26.ffn_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.27.attention.dyn_w_proj.dd": "pytorch_model-00003-of-00003.bin", "layers.27.attention.dyn_w_proj.dw1": "pytorch_model-00003-of-00003.bin", "layers.27.attention.dyn_w_proj.dw_m": "pytorch_model-00003-of-00003.bin", "layers.27.attention.dyn_w_proj.post_proj.w": "pytorch_model-00003-of-00003.bin", "layers.27.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00003-of-00003.bin", "layers.27.attention.dyn_w_proj.qkw": "pytorch_model-00003-of-00003.bin", "layers.27.attention.dyn_w_proj.qkw_m": "pytorch_model-00003-of-00003.bin", "layers.27.attention.k_norm.scale": "pytorch_model-00003-of-00003.bin", "layers.27.attention.kv_cache.k_cache": "pytorch_model-00003-of-00003.bin", "layers.27.attention.kv_cache.kw_cache": "pytorch_model-00003-of-00003.bin", "layers.27.attention.kv_cache.v_cache": "pytorch_model-00003-of-00003.bin", "layers.27.attention.q_norm.scale": "pytorch_model-00003-of-00003.bin", "layers.27.attention.wo.bias": "pytorch_model-00003-of-00003.bin", "layers.27.attention.wo.weight": "pytorch_model-00003-of-00003.bin", "layers.27.attention.wqkv.bias": "pytorch_model-00003-of-00003.bin", "layers.27.attention.wqkv.weight": "pytorch_model-00003-of-00003.bin", "layers.27.attention_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.27.attention_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.27.feed_forward.w1.bias": "pytorch_model-00003-of-00003.bin", "layers.27.feed_forward.w1.weight": "pytorch_model-00003-of-00003.bin", "layers.27.feed_forward.w2.bias": "pytorch_model-00003-of-00003.bin", "layers.27.feed_forward.w2.weight": "pytorch_model-00003-of-00003.bin", "layers.27.ffn_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.27.ffn_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.28.attention.dyn_w_proj.dd": "pytorch_model-00003-of-00003.bin", "layers.28.attention.dyn_w_proj.dw1": "pytorch_model-00003-of-00003.bin", "layers.28.attention.dyn_w_proj.dw_m": "pytorch_model-00003-of-00003.bin", "layers.28.attention.dyn_w_proj.post_proj.w": "pytorch_model-00003-of-00003.bin", "layers.28.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00003-of-00003.bin", "layers.28.attention.dyn_w_proj.qkw": "pytorch_model-00003-of-00003.bin", "layers.28.attention.dyn_w_proj.qkw_m": "pytorch_model-00003-of-00003.bin", "layers.28.attention.k_norm.scale": "pytorch_model-00003-of-00003.bin", "layers.28.attention.kv_cache.k_cache": "pytorch_model-00003-of-00003.bin", "layers.28.attention.kv_cache.kw_cache": "pytorch_model-00003-of-00003.bin", "layers.28.attention.kv_cache.v_cache": "pytorch_model-00003-of-00003.bin", "layers.28.attention.q_norm.scale": "pytorch_model-00003-of-00003.bin", "layers.28.attention.wo.bias": "pytorch_model-00003-of-00003.bin", "layers.28.attention.wo.weight": "pytorch_model-00003-of-00003.bin", "layers.28.attention.wqkv.bias": "pytorch_model-00003-of-00003.bin", "layers.28.attention.wqkv.weight": "pytorch_model-00003-of-00003.bin", "layers.28.attention_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.28.attention_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.28.feed_forward.w1.bias": "pytorch_model-00003-of-00003.bin", "layers.28.feed_forward.w1.weight": "pytorch_model-00003-of-00003.bin", "layers.28.feed_forward.w2.bias": "pytorch_model-00003-of-00003.bin", "layers.28.feed_forward.w2.weight": "pytorch_model-00003-of-00003.bin", "layers.28.ffn_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.28.ffn_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.29.attention.dyn_w_proj.dd": "pytorch_model-00003-of-00003.bin", "layers.29.attention.dyn_w_proj.dw1": "pytorch_model-00003-of-00003.bin", "layers.29.attention.dyn_w_proj.dw_m": "pytorch_model-00003-of-00003.bin", "layers.29.attention.dyn_w_proj.post_proj.w": "pytorch_model-00003-of-00003.bin", "layers.29.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00003-of-00003.bin", "layers.29.attention.dyn_w_proj.qkw": "pytorch_model-00003-of-00003.bin", "layers.29.attention.dyn_w_proj.qkw_m": "pytorch_model-00003-of-00003.bin", "layers.29.attention.k_norm.scale": "pytorch_model-00003-of-00003.bin", "layers.29.attention.kv_cache.k_cache": "pytorch_model-00003-of-00003.bin", "layers.29.attention.kv_cache.kw_cache": "pytorch_model-00003-of-00003.bin", "layers.29.attention.kv_cache.v_cache": "pytorch_model-00003-of-00003.bin", "layers.29.attention.q_norm.scale": "pytorch_model-00003-of-00003.bin", "layers.29.attention.wo.bias": "pytorch_model-00003-of-00003.bin", "layers.29.attention.wo.weight": "pytorch_model-00003-of-00003.bin", "layers.29.attention.wqkv.bias": "pytorch_model-00003-of-00003.bin", "layers.29.attention.wqkv.weight": "pytorch_model-00003-of-00003.bin", "layers.29.attention_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.29.attention_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.29.feed_forward.w1.bias": "pytorch_model-00003-of-00003.bin", "layers.29.feed_forward.w1.weight": "pytorch_model-00003-of-00003.bin", "layers.29.feed_forward.w2.bias": "pytorch_model-00003-of-00003.bin", "layers.29.feed_forward.w2.weight": "pytorch_model-00003-of-00003.bin", "layers.29.ffn_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.29.ffn_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.3.attention.dyn_w_proj.dd": "pytorch_model-00001-of-00003.bin", "layers.3.attention.dyn_w_proj.dw1": "pytorch_model-00001-of-00003.bin", "layers.3.attention.dyn_w_proj.dw_m": "pytorch_model-00001-of-00003.bin", "layers.3.attention.dyn_w_proj.post_proj.w": "pytorch_model-00001-of-00003.bin", "layers.3.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00001-of-00003.bin", "layers.3.attention.dyn_w_proj.qkw": "pytorch_model-00001-of-00003.bin", "layers.3.attention.dyn_w_proj.qkw_m": "pytorch_model-00001-of-00003.bin", "layers.3.attention.k_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.3.attention.kv_cache.k_cache": "pytorch_model-00001-of-00003.bin", "layers.3.attention.kv_cache.kw_cache": "pytorch_model-00001-of-00003.bin", "layers.3.attention.kv_cache.v_cache": "pytorch_model-00001-of-00003.bin", "layers.3.attention.q_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.3.attention.wo.bias": "pytorch_model-00001-of-00003.bin", "layers.3.attention.wo.weight": "pytorch_model-00001-of-00003.bin", "layers.3.attention.wqkv.bias": "pytorch_model-00001-of-00003.bin", "layers.3.attention.wqkv.weight": "pytorch_model-00001-of-00003.bin", "layers.3.attention_norm.bias": "pytorch_model-00001-of-00003.bin", "layers.3.attention_norm.weight": "pytorch_model-00001-of-00003.bin", "layers.3.feed_forward.w1.bias": "pytorch_model-00001-of-00003.bin", "layers.3.feed_forward.w1.weight": "pytorch_model-00001-of-00003.bin", "layers.3.feed_forward.w2.bias": "pytorch_model-00001-of-00003.bin", "layers.3.feed_forward.w2.weight": "pytorch_model-00001-of-00003.bin", "layers.3.ffn_norm.bias": "pytorch_model-00001-of-00003.bin", "layers.3.ffn_norm.weight": "pytorch_model-00001-of-00003.bin", "layers.30.attention.dyn_w_proj.dd": "pytorch_model-00003-of-00003.bin", "layers.30.attention.dyn_w_proj.dw1": "pytorch_model-00003-of-00003.bin", "layers.30.attention.dyn_w_proj.dw_m": "pytorch_model-00003-of-00003.bin", "layers.30.attention.dyn_w_proj.post_proj.w": "pytorch_model-00003-of-00003.bin", "layers.30.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00003-of-00003.bin", "layers.30.attention.dyn_w_proj.qkw": "pytorch_model-00003-of-00003.bin", "layers.30.attention.dyn_w_proj.qkw_m": "pytorch_model-00003-of-00003.bin", "layers.30.attention.k_norm.scale": "pytorch_model-00003-of-00003.bin", "layers.30.attention.kv_cache.k_cache": "pytorch_model-00003-of-00003.bin", "layers.30.attention.kv_cache.kw_cache": "pytorch_model-00003-of-00003.bin", "layers.30.attention.kv_cache.v_cache": "pytorch_model-00003-of-00003.bin", "layers.30.attention.q_norm.scale": "pytorch_model-00003-of-00003.bin", "layers.30.attention.wo.bias": "pytorch_model-00003-of-00003.bin", "layers.30.attention.wo.weight": "pytorch_model-00003-of-00003.bin", "layers.30.attention.wqkv.bias": "pytorch_model-00003-of-00003.bin", "layers.30.attention.wqkv.weight": "pytorch_model-00003-of-00003.bin", "layers.30.attention_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.30.attention_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.30.feed_forward.w1.bias": "pytorch_model-00003-of-00003.bin", "layers.30.feed_forward.w1.weight": "pytorch_model-00003-of-00003.bin", "layers.30.feed_forward.w2.bias": "pytorch_model-00003-of-00003.bin", "layers.30.feed_forward.w2.weight": "pytorch_model-00003-of-00003.bin", "layers.30.ffn_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.30.ffn_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.31.attention.dyn_w_proj.dd": "pytorch_model-00003-of-00003.bin", "layers.31.attention.dyn_w_proj.dw1": "pytorch_model-00003-of-00003.bin", "layers.31.attention.dyn_w_proj.dw_m": "pytorch_model-00003-of-00003.bin", "layers.31.attention.dyn_w_proj.post_proj.w": "pytorch_model-00003-of-00003.bin", "layers.31.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00003-of-00003.bin", "layers.31.attention.dyn_w_proj.qkw": "pytorch_model-00003-of-00003.bin", "layers.31.attention.dyn_w_proj.qkw_m": "pytorch_model-00003-of-00003.bin", "layers.31.attention.k_norm.scale": "pytorch_model-00003-of-00003.bin", "layers.31.attention.kv_cache.k_cache": "pytorch_model-00003-of-00003.bin", "layers.31.attention.kv_cache.kw_cache": "pytorch_model-00003-of-00003.bin", "layers.31.attention.kv_cache.v_cache": "pytorch_model-00003-of-00003.bin", "layers.31.attention.q_norm.scale": "pytorch_model-00003-of-00003.bin", "layers.31.attention.wo.bias": "pytorch_model-00003-of-00003.bin", "layers.31.attention.wo.weight": "pytorch_model-00003-of-00003.bin", "layers.31.attention.wqkv.bias": "pytorch_model-00003-of-00003.bin", "layers.31.attention.wqkv.weight": "pytorch_model-00003-of-00003.bin", "layers.31.attention_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.31.attention_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.31.feed_forward.w1.bias": "pytorch_model-00003-of-00003.bin", "layers.31.feed_forward.w1.weight": "pytorch_model-00003-of-00003.bin", "layers.31.feed_forward.w2.bias": "pytorch_model-00003-of-00003.bin", "layers.31.feed_forward.w2.weight": "pytorch_model-00003-of-00003.bin", "layers.31.ffn_norm.bias": "pytorch_model-00003-of-00003.bin", "layers.31.ffn_norm.weight": "pytorch_model-00003-of-00003.bin", "layers.4.attention.dyn_w_proj.dd": "pytorch_model-00001-of-00003.bin", "layers.4.attention.dyn_w_proj.dw1": "pytorch_model-00001-of-00003.bin", "layers.4.attention.dyn_w_proj.dw_m": "pytorch_model-00001-of-00003.bin", "layers.4.attention.dyn_w_proj.post_proj.w": "pytorch_model-00001-of-00003.bin", "layers.4.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00001-of-00003.bin", "layers.4.attention.dyn_w_proj.qkw": "pytorch_model-00001-of-00003.bin", "layers.4.attention.dyn_w_proj.qkw_m": "pytorch_model-00001-of-00003.bin", "layers.4.attention.k_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.4.attention.kv_cache.k_cache": "pytorch_model-00001-of-00003.bin", "layers.4.attention.kv_cache.kw_cache": "pytorch_model-00001-of-00003.bin", "layers.4.attention.kv_cache.v_cache": "pytorch_model-00001-of-00003.bin", "layers.4.attention.q_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.4.attention.wo.bias": "pytorch_model-00001-of-00003.bin", "layers.4.attention.wo.weight": "pytorch_model-00001-of-00003.bin", "layers.4.attention.wqkv.bias": "pytorch_model-00001-of-00003.bin", "layers.4.attention.wqkv.weight": "pytorch_model-00001-of-00003.bin", "layers.4.attention_norm.bias": "pytorch_model-00001-of-00003.bin", "layers.4.attention_norm.weight": "pytorch_model-00001-of-00003.bin", "layers.4.feed_forward.w1.bias": "pytorch_model-00001-of-00003.bin", "layers.4.feed_forward.w1.weight": "pytorch_model-00001-of-00003.bin", "layers.4.feed_forward.w2.bias": "pytorch_model-00001-of-00003.bin", "layers.4.feed_forward.w2.weight": "pytorch_model-00001-of-00003.bin", "layers.4.ffn_norm.bias": "pytorch_model-00001-of-00003.bin", "layers.4.ffn_norm.weight": "pytorch_model-00001-of-00003.bin", "layers.5.attention.dyn_w_proj.dd": "pytorch_model-00001-of-00003.bin", "layers.5.attention.dyn_w_proj.dw1": "pytorch_model-00001-of-00003.bin", "layers.5.attention.dyn_w_proj.dw_m": "pytorch_model-00001-of-00003.bin", "layers.5.attention.dyn_w_proj.post_proj.w": "pytorch_model-00001-of-00003.bin", "layers.5.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00001-of-00003.bin", "layers.5.attention.dyn_w_proj.qkw": "pytorch_model-00001-of-00003.bin", "layers.5.attention.dyn_w_proj.qkw_m": "pytorch_model-00001-of-00003.bin", "layers.5.attention.k_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.5.attention.kv_cache.k_cache": "pytorch_model-00001-of-00003.bin", "layers.5.attention.kv_cache.kw_cache": "pytorch_model-00001-of-00003.bin", "layers.5.attention.kv_cache.v_cache": "pytorch_model-00001-of-00003.bin", "layers.5.attention.q_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.5.attention.wo.bias": "pytorch_model-00001-of-00003.bin", "layers.5.attention.wo.weight": "pytorch_model-00001-of-00003.bin", "layers.5.attention.wqkv.bias": "pytorch_model-00001-of-00003.bin", "layers.5.attention.wqkv.weight": "pytorch_model-00001-of-00003.bin", "layers.5.attention_norm.bias": "pytorch_model-00001-of-00003.bin", "layers.5.attention_norm.weight": "pytorch_model-00001-of-00003.bin", "layers.5.feed_forward.w1.bias": "pytorch_model-00001-of-00003.bin", "layers.5.feed_forward.w1.weight": "pytorch_model-00001-of-00003.bin", "layers.5.feed_forward.w2.bias": "pytorch_model-00001-of-00003.bin", "layers.5.feed_forward.w2.weight": "pytorch_model-00001-of-00003.bin", "layers.5.ffn_norm.bias": "pytorch_model-00001-of-00003.bin", "layers.5.ffn_norm.weight": "pytorch_model-00001-of-00003.bin", "layers.6.attention.dyn_w_proj.dd": "pytorch_model-00001-of-00003.bin", "layers.6.attention.dyn_w_proj.dw1": "pytorch_model-00001-of-00003.bin", "layers.6.attention.dyn_w_proj.dw_m": "pytorch_model-00001-of-00003.bin", "layers.6.attention.dyn_w_proj.post_proj.w": "pytorch_model-00001-of-00003.bin", "layers.6.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00001-of-00003.bin", "layers.6.attention.dyn_w_proj.qkw": "pytorch_model-00001-of-00003.bin", "layers.6.attention.dyn_w_proj.qkw_m": "pytorch_model-00001-of-00003.bin", "layers.6.attention.k_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.6.attention.kv_cache.k_cache": "pytorch_model-00001-of-00003.bin", "layers.6.attention.kv_cache.kw_cache": "pytorch_model-00001-of-00003.bin", "layers.6.attention.kv_cache.v_cache": "pytorch_model-00001-of-00003.bin", "layers.6.attention.q_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.6.attention.wo.bias": "pytorch_model-00001-of-00003.bin", "layers.6.attention.wo.weight": "pytorch_model-00001-of-00003.bin", "layers.6.attention.wqkv.bias": "pytorch_model-00001-of-00003.bin", "layers.6.attention.wqkv.weight": "pytorch_model-00001-of-00003.bin", "layers.6.attention_norm.bias": "pytorch_model-00001-of-00003.bin", "layers.6.attention_norm.weight": "pytorch_model-00001-of-00003.bin", "layers.6.feed_forward.w1.bias": "pytorch_model-00001-of-00003.bin", "layers.6.feed_forward.w1.weight": "pytorch_model-00001-of-00003.bin", "layers.6.feed_forward.w2.bias": "pytorch_model-00001-of-00003.bin", "layers.6.feed_forward.w2.weight": "pytorch_model-00001-of-00003.bin", "layers.6.ffn_norm.bias": "pytorch_model-00001-of-00003.bin", "layers.6.ffn_norm.weight": "pytorch_model-00001-of-00003.bin", "layers.7.attention.dyn_w_proj.dd": "pytorch_model-00001-of-00003.bin", "layers.7.attention.dyn_w_proj.dw1": "pytorch_model-00001-of-00003.bin", "layers.7.attention.dyn_w_proj.dw_m": "pytorch_model-00001-of-00003.bin", "layers.7.attention.dyn_w_proj.post_proj.w": "pytorch_model-00001-of-00003.bin", "layers.7.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00001-of-00003.bin", "layers.7.attention.dyn_w_proj.qkw": "pytorch_model-00001-of-00003.bin", "layers.7.attention.dyn_w_proj.qkw_m": "pytorch_model-00001-of-00003.bin", "layers.7.attention.k_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.7.attention.kv_cache.k_cache": "pytorch_model-00001-of-00003.bin", "layers.7.attention.kv_cache.kw_cache": "pytorch_model-00001-of-00003.bin", "layers.7.attention.kv_cache.v_cache": "pytorch_model-00001-of-00003.bin", "layers.7.attention.q_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.7.attention.wo.bias": "pytorch_model-00001-of-00003.bin", "layers.7.attention.wo.weight": "pytorch_model-00001-of-00003.bin", "layers.7.attention.wqkv.bias": "pytorch_model-00001-of-00003.bin", "layers.7.attention.wqkv.weight": "pytorch_model-00001-of-00003.bin", "layers.7.attention_norm.bias": "pytorch_model-00001-of-00003.bin", "layers.7.attention_norm.weight": "pytorch_model-00001-of-00003.bin", "layers.7.feed_forward.w1.bias": "pytorch_model-00001-of-00003.bin", "layers.7.feed_forward.w1.weight": "pytorch_model-00001-of-00003.bin", "layers.7.feed_forward.w2.bias": "pytorch_model-00001-of-00003.bin", "layers.7.feed_forward.w2.weight": "pytorch_model-00001-of-00003.bin", "layers.7.ffn_norm.bias": "pytorch_model-00001-of-00003.bin", "layers.7.ffn_norm.weight": "pytorch_model-00001-of-00003.bin", "layers.8.attention.dyn_w_proj.dd": "pytorch_model-00001-of-00003.bin", "layers.8.attention.dyn_w_proj.dw1": "pytorch_model-00001-of-00003.bin", "layers.8.attention.dyn_w_proj.dw_m": "pytorch_model-00001-of-00003.bin", "layers.8.attention.dyn_w_proj.post_proj.w": "pytorch_model-00001-of-00003.bin", "layers.8.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00001-of-00003.bin", "layers.8.attention.dyn_w_proj.qkw": "pytorch_model-00001-of-00003.bin", "layers.8.attention.dyn_w_proj.qkw_m": "pytorch_model-00001-of-00003.bin", "layers.8.attention.k_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.8.attention.kv_cache.k_cache": "pytorch_model-00001-of-00003.bin", "layers.8.attention.kv_cache.kw_cache": "pytorch_model-00001-of-00003.bin", "layers.8.attention.kv_cache.v_cache": "pytorch_model-00001-of-00003.bin", "layers.8.attention.q_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.8.attention.wo.bias": "pytorch_model-00001-of-00003.bin", "layers.8.attention.wo.weight": "pytorch_model-00001-of-00003.bin", "layers.8.attention.wqkv.bias": "pytorch_model-00001-of-00003.bin", "layers.8.attention.wqkv.weight": "pytorch_model-00001-of-00003.bin", "layers.8.attention_norm.bias": "pytorch_model-00001-of-00003.bin", "layers.8.attention_norm.weight": "pytorch_model-00001-of-00003.bin", "layers.8.feed_forward.w1.bias": "pytorch_model-00001-of-00003.bin", "layers.8.feed_forward.w1.weight": "pytorch_model-00001-of-00003.bin", "layers.8.feed_forward.w2.bias": "pytorch_model-00001-of-00003.bin", "layers.8.feed_forward.w2.weight": "pytorch_model-00001-of-00003.bin", "layers.8.ffn_norm.bias": "pytorch_model-00001-of-00003.bin", "layers.8.ffn_norm.weight": "pytorch_model-00001-of-00003.bin", "layers.9.attention.dyn_w_proj.dd": "pytorch_model-00001-of-00003.bin", "layers.9.attention.dyn_w_proj.dw1": "pytorch_model-00001-of-00003.bin", "layers.9.attention.dyn_w_proj.dw_m": "pytorch_model-00001-of-00003.bin", "layers.9.attention.dyn_w_proj.post_proj.w": "pytorch_model-00001-of-00003.bin", "layers.9.attention.dyn_w_proj.pre_proj.w": "pytorch_model-00001-of-00003.bin", "layers.9.attention.dyn_w_proj.qkw": "pytorch_model-00001-of-00003.bin", "layers.9.attention.dyn_w_proj.qkw_m": "pytorch_model-00001-of-00003.bin", "layers.9.attention.k_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.9.attention.kv_cache.k_cache": "pytorch_model-00001-of-00003.bin", "layers.9.attention.kv_cache.kw_cache": "pytorch_model-00001-of-00003.bin", "layers.9.attention.kv_cache.v_cache": "pytorch_model-00001-of-00003.bin", "layers.9.attention.q_norm.scale": "pytorch_model-00001-of-00003.bin", "layers.9.attention.wo.bias": "pytorch_model-00001-of-00003.bin", "layers.9.attention.wo.weight": "pytorch_model-00001-of-00003.bin", "layers.9.attention.wqkv.bias": "pytorch_model-00001-of-00003.bin", "layers.9.attention.wqkv.weight": "pytorch_model-00001-of-00003.bin", "layers.9.attention_norm.bias": "pytorch_model-00001-of-00003.bin", "layers.9.attention_norm.weight": "pytorch_model-00001-of-00003.bin", "layers.9.feed_forward.w1.bias": "pytorch_model-00001-of-00003.bin", "layers.9.feed_forward.w1.weight": "pytorch_model-00001-of-00003.bin", "layers.9.feed_forward.w2.bias": "pytorch_model-00001-of-00003.bin", "layers.9.feed_forward.w2.weight": "pytorch_model-00001-of-00003.bin", "layers.9.ffn_norm.bias": "pytorch_model-00001-of-00003.bin", "layers.9.ffn_norm.weight": "pytorch_model-00001-of-00003.bin", "norm.bias": "pytorch_model-00003-of-00003.bin", "norm.weight": "pytorch_model-00003-of-00003.bin", "output.weight": "pytorch_model-00003-of-00003.bin", "tok_embeddings.weight": "pytorch_model-00001-of-00003.bin" } }